Functional Characterization of the Ocular Prostaglandin F2α (PGF2α) Receptor

Prostaglandin F2α(PGF2α) is a product of cyclooxygenase-catalyzed metabolism of arachidonic acid. Recently, PGF2α analogs have been hypothesized to reduce intraocular pressure via relaxation of the ciliary muscle. To investigate the molecular basis of PGF2α receptor (FP) activation in the eye, we cloned the FP from a human ciliary body (hcb) cDNA library. The open reading frame of the hcb-FP cDNA was identical to the uterine FP cDNA. The hcb-FP appeared to be predominantly membrane-localized, as visualized by an FP-specific peptide antibody, and coupled to inositol phosphate formation when stably expressed in HEK 293 cells. Interestingly, the hcb-FP could also be activated by the F2isoprostane, 12-iso-PGF2α, in addition to its cognate ligand, PGF2α. 12-iso-PGF2α was less potent (EC50 = 5 μm) than PGF2α(EC50 = 10 nm) in generating inositol phosphates via the hcb-FP in HEK 293 cells. Both ligands also stimulated mitogenesis in NIH 3T3 cells. Although 12-iso-PGF2α caused a dose-dependent activation of the FP, it failed to activate the recombinant human prostacyclin receptor and caused only minimal activation of the thromboxane receptor isoforms stably expressed in HEK 293 cells. Four additional F2 isoprostanes, 8-iso-PGF2α, IPF2α-I, IPF2α-III, and 9β,11β-PGF2, caused trivial, or no, activation of the FP. Consistent with these observations, only PGF2α and 12-iso-PGF2α caused rapid homologous desensitization of FP and also exhibited cross-desensitization, with PGF2α resulting in a maximum of ∼60% desensitization. The human FP may thus be activated specifically, by the free radical-catalyzed F2 isoprostane, 12-iso-PGF2α, in addition to the cyclooxygenase product, PGF2α. Incidental receptor activation by isoprostanes may complement the actions of PGF2α in clinical syndromes where oxidant stress and augmented prostaglandin biosynthesis coincide.

Recently, PGF 2␣ has also been shown to cause hypertrophy of cardiac myocytes and induction of myofibrillar genes, independent of muscle contraction. These observations suggest a role for the eicosanoid during development, in compensatory hypertrophy and/or in recovery of the heart from injury (12).
Recently, PGF 2␣ analogs have been shown to reduce intraocular pressure (IOP), in patients with glaucoma (13,14). Although the precise mechanisms involved remain unclear, the effects of PGF 2␣ analogs on IOP may be attributed, at least in part, to their actions on the ciliary muscle. PGF 2␣ reduces IOP by increasing the uveoscleral outflow of aqueous humor (15,16), possibly by reducing the resistance between the ciliary muscle bundles, via an effect on the extracellular matrix (17).
A single PGF 2␣ receptor (FP) has been cloned from myometrial tissue (18 -22). Given that there is evidence consistent with splice variation of the FP (23), as has been described for other prostanoid receptors (24,25), we wished to address the possibility that a distinct isoform might mediate the actions of PGF 2␣ in the ciliary muscle. Clarification of the nature of the human ciliary FP and development of an antibody that specifically recognized the receptor protein would facilitate investigation of the effects of PGF 2␣ and its analogs on IOP.
PGF 2␣ is formed from arachidonic acid via metabolic transformation sequentially catalyzed by phospholipases, cyclooxygenases, and a specific PGF synthase (26). However, it is now appreciated that a series of PGF 2␣ isomers, the F 2 isoprostanes, may also be formed in vivo via a free radical-dependent pathway (27)(28)(29). It has been speculated that these F 2 isoprostanes may function as incidental ligands at eicosanoid receptors, and, possibly, activate related receptors of their own (30). To date, attention has focused particularly on 8-iso-PGF 2␣ . This compound is a potent vasoconstrictor. It is also a mitogen and may activate human platelets (31)(32)(33). Curiously, despite its F prostaglandin configuration, 8-iso-PGF 2␣ has been shown to activate thromboxane receptors (TPs), and its biological effects are blocked by TP antagonists (31)(32)(33).
We now report the cloning of an FP receptor from the human ciliary body (hcb) cDNA library and its localization on the cell membrane. The gene product is identical to that cloned from * This work was supported in part by National Institutes of Health Grants HL54500, HL07843, HL57847 (to G. A. F.), and DK44730 (to J. R.) and by National Science Foundation Grant CHE-9013145. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Cloning of the Human Ocular FP Receptor cDNA-Phage DNA was prepared from the hcb-cDNA library (kindly donated by Dr. Miguel CocaPrados, Yale University, New Haven, CT) and subjected to PCR. PCR was performed on 200 ng of human ciliary body cDNA with 100 pmol of the FP-specific sense primer 5Ј-TCGAGGACCTGGTGTTTC-TAC-3Ј (18) and a degenerate antisense primer 5Ј-CCAIGGRTCIARDA-TYTGRTT-3Ј (I ϭ inosine, R ϭ G/A, D ϭ G/A/T, Y ϭ T/C), 1 ϫ Ampli-Taq buffer, 3 mM MgCl 2 , and 0.5 mM dNTPs in a total reaction volume of 100 l. The samples were subjected to a "Hot Start" as described previously (37), followed by the addition of 0.5 units of Ampli-Taq DNA polymerase. The reaction was then subjected to denaturation at 99°C for 1 min, annealing at 50°C for 2 min, and extension at 72°C for 3 min for 5 cycles, followed by denaturation at 99°C for 1 mine, annealing at 55°C for 2 min, and extension at 72°C for 3 min for 25 cycles. PCR products electrophoresed on a 1% agarose gel revealed the presence of a ϳ369-bp band, which generated a positive signal when subjected to Southern blot hybridization with a 32 P-labeled FP-specific oligonucleotide, 5Ј-GACTGGGAAGATAGATTTTAT-3Ј (18). The ϳ360-bp PCR product was subcloned into pBluescript to isolate the full-length hcb-FP cDNA and was then used as a probe to screen the hcb-cDNA library in -Uni Zap-XR, as described earlier (37). Hybridization was performed in Rapid-hyb buffer at 65°C for 3 h. Two positive clones were identified, isolated, rescued, and sequenced. The full-length hcb-FP cDNA isolated was ϳ3.1 kilobase pairs in size, consisting of 151 bp of 5Ј-untranslated region, 1077 bp of open reading frame, and 1867 bp of 3Ј-untranslated region, ending in a poly(A) tail. The open reading frame of the hcb-FP encodes a 359-amino acid protein with seven putative membranespanning domains, belonging to the superfamily of G protein-coupled receptors.
Northern Blot Analysis-Tissue distribution of the human FP mRNA was analyzed on human multiple tissue Northern blots from CLON-TECH (Palo Alto, CA) using a ϳ650-bp BamHI/HindIII fragment of the full-length FP clone random primed with [ 32 P]dCTP to a specific activity of 3.1 ϫ 10 9 cpm/g. The blots were hybridized in Rapid-hyb buffer at 65°C for 3 h and washed initially with 50 ml of 5 ϫ SSC, 0.1% SDS at room temperature for 30 min and then with four washes of 50 ml of 0.2 ϫ SSC, 0.1% SDS at 60°C for 30 min. The blots were then autoradiographed overnight at Ϫ80°C.
Stable Expression in HEK 293 Cells-A ϳ 1.8-kilobase pair EcoRI fragment from the full-length hcb-FP cDNA was subcloned into pcDNA3 (Invitrogen, San Diego, CA). The orientation of the insert was verified by restriction digestions. This expression construct (pcDNA3-FP) was then used to transfect HEK 293 cells using Dotap under standard conditions. HEK 293 cells were also transfected with pcDNA3 to serve as a control. The medium was replaced after 6 h with fresh medium containing 1.5 mg/ml G418. Stable transfectants were selected on medium containing 1.5 mg/ml G418 and screened for the expression of the FP by binding to [ 3 H]PGF 2␣ and second messenger (inositol phosphate; InsP) generation. One clone (HEK-FP), out of 19 clones selected, was chosen for further characterization.

Measurement of [ 3 H]Inositol
Phosphate Formation-To study the signal transduction properties of the hcb-FP, confluent cultures of HEK-FP cells in 12-well plates were labeled to equilibrium with myo-[2-3 H]inositol (2 Ci/ml) for 16 -24 h in serum-free DMEM containing 20 mM HEPES, pH 7.5, and 0.5% Albumax. Cells were preincubated in this medium with 20 mM LiCl for 15 min at 37°C and then stimulated directly by the addition of agonist for 5-10 min at 37°C. Total InsP formation was measured as described previously (38). Briefly, InsP formation was stopped by aspiration of the medium, addition of 0.75 ml of 10 mM formic acid and incubation at room temperature for 30 min. The solution containing the extracted InsP was neutralized and diluted with 3 ml of 10 mM NH 4 OH (yielding a final pH of 8 -9) and then applied directly to a column containing 0.7 ml of the anion exchange resin, AG 1-X8. The column was washed with 4 ml of 40 mM ammonium formate, pH 5.0, to remove the free inositol and the glyceroinositol. Total InsPs were eluted with 4 ml of 2 M ammonium formate, pH 5.0. One ml of the eluate was counted with 9 ml of scintillation fluid. Results presented are an average of three to five independent experiments.
Desensitization experiments were performed essentially as described by Opperman et al. (39). Briefly, after incubation with 20 mM LiCl, cells were pretreated with PGF 2␣ , 12-iso-PGF 2␣ , or PBS (control) for 5 min at 37°C, followed by immediate aspiration of the medium. The cells were then washed twice with 1 ml of 50 mM glycine, 150 mM NaCl, pH 3.0. The cells were then restimulated with PGF 2␣ or 12-iso-PGF 2␣ in medium containing 20 mM LiCl for 10 min at 37°C. The reactions were terminated and InsPs were extracted as described above. Results presented are an average of three independent experiments.
Generation of FP Antibodies-Polyclonal peptide antibodies were raised in rabbits to the sequence GINGNHSLETCET corresponding to the third extracellular loop of the human FP receptor by Research Genetics Inc (Huntsville, AL). The antisera were tested by immunoblotting, using membranes from HEK-FP cells. HEK-FP membranes were prepared from confluent 100-mm dishes as follows. Briefly, cells were washed once with PBS and scrapped into 20 mM Tris, pH 7.4, containing 4 mM EDTA, 10 g/ml aprotinin, 10 g/ml leupeptin, and 0.2 mM phenylmethylsulfonyl fluoride. Cells were lysed by sonication on ice, and membrane fractions were collected by centrifugation at 115,000 ϫ g for 1 h at 4°C. The resulting pellet was resuspended in the same buffer. Membrane proteins (100 g/lane) were resolved on a 10% SDS-polyacrylamide gel and transferred to a nitrocellulose membrane. The immunoblot was first blocked with 5% milk in TBS-T. FPs were visualized by treating the immunoblots with 1:500 dilution of the crude peptide antisera (in 5% milk/TBS-T) for 1 h at room temperature, followed by horseradish peroxidase-conjugated anti-rabbit IgG (1:5000 dilution). Antigen-antibody complexes were visualized by chemiluminescence.
For immunocytochemistry, cells were grown on chamber slides (Nunc, Napierville, IL) and fixed with 70% methanol, 30% acetone at Ϫ20°C for 10 min, followed by incubation at room temperature for 5 min. The cells were blocked with 2% BSA/PBS and then treated with 1:200 dilution of anti-FP antisera in 0.5% BSA/PBS for 1 h at room temperature, followed by a 1-h incubation with fluorescein isothiocyanate-labeled anti-rabbit IgG (1:500) in 0.5% BSA/PBS. Between each step, slides were washed three times for 10 min each with PBS. Slides were mounted in Vectashield (Vector Laboratories, Burlingame, CA) and examined by fluorescence microscopy with a Nikon Microphot FXA microscope.
Assay for DNA Synthesis-We measured [methyl-3 H]thymidine incorporation into DNA by the method of Nakamura et al. (40) with slight modifications. NIH 3T3 cells were subcultured into 12-well plates.

RESULTS
The hcb-FP encodes an open reading frame of 359 amino acids that is identical to the uterine FP. Northern blot analysis reveals the FP mRNA to be ϳ5 kilobase pairs in size and highly expressed in the human heart Ͼ pancreas Ͼ liver, placenta Ͼ skeletal muscle Ͼ uterus Ͼ kidney Ͼ small intestine (Fig. 1).
We generated a mammalian expression construct of the hcb-FP cDNA in pcDNA3 and used this to transfect HEK 293 cells. One of the stable transfectants (HEK-FP), was chosen for detailed characterization.
Activation of the FP by PGF 2␣ leads to an increase in InsP formation in HEK 293 cells in a dose-dependent manner, reaching a plateau around 1 M PGF 2␣ . The EC 50 for PGF 2␣ -induced InsP formation is 10 Ϯ 1.5 nM (Fig. 2). F 2 -isoprostanes are isomers of PGF 2␣ and have been divided into four structural classes (41)(42)(43). To test their possible affinity for the FP, we selected members of the first (IPF 2␣ -I), third (IPF 2␣ -III), and fourth classes (8-iso-PGF 2␣ , 12-iso-PGF 2␣ , and 9␤, 11␤-PGF 2 ) (Fig. 3). As seen in Fig. 4, only 12-iso-PGF 2␣ , among these isoprostanes, caused significant activation of the hcb-FP, as observed by InsP formation in HEK-FP cells. To explore the specificity of 12-iso-PGF 2␣ for the FP, these compounds were also tested for their ability to activate other prostanoid receptors, namely, the IP and the two cloned isoforms of the thromboxane receptor (TP␣ and TP␤). Iloprost (a prostacyclin analog) induced a ϳ2.6 Ϯ 0.4-fold increase in InsP formation in HEK 293 cells stably expressing the human IP receptor (44). However, none of the isoprostanes mimicked this response. On the other hand, when tested on HEK 293 cells stably expressing the TP receptor isoforms (45), 12-iso-PGF 2␣ resulted in 1.6 -1.8-fold increase in InsP formation (Fig. 4). In comparison, equimolar concentrations of U46619, a thromboxane agonist, caused 9 -10-fold stimulation of InsP formation in these cells. The 12-iso-PGF 2␣ -induced InsP formation via the TP was abolished by the TP antagonist, SQ29548. As demon-strated previously (33), 8-iso-PGF 2␣ also activates TPs, and this response was also abolished by SQ29548.
We generated polyclonal peptide antibodies to the third extracellular loop of the FP. Immunoblot analysis of HEK-FP membranes with anti-FP antisera revealed FP to be a broad complex with a molecular weight ranging from 42 to 55 kDa (Fig. 6A). This signal appeared to be specific to FP inasmuch as it was not evident in pcDNA3 vector-transfected HEK 293 cells. Furthermore, the FP signal was competed away by preincubat- FIG. 1. Tissue distribution of human FP. Northern blot analysis of 2 g of human mRNA from heart, brain, placenta, lung, liver, skeletal muscle, kidney, pancreas, spleen, thymus, prostate, testis, uterus, small intestine, colon, and peripheral blood leukocytes. The multiple tissue Northern blot was hybridized with a ϳ650-bp BamHI/HindIII fragment from the hcb-FP cDNA as described under "Experimental Procedures." The blots were washed with 0.2% SSC, 0.1% SDS at 60°C, followed by autoradiography. ing the antisera with 10 g/ml of the corresponding peptide. The human anti-FP antiserum also recognizes the native FP in mouse NIH 3T3 cells. Immunocytochemistry of HEK-FP cells reveals FP to be expressed predominantly at the cell surface (Fig. 6B).
We investigated the ability of FP to undergo agonist-induced rapid homologous desensitization. Although receptor desensitization is a common method of regulation among G proteincoupled receptors (46), there is very little information available on the regulation of FP function by desensitization. Pretreatment of HEK-FP cells with 1 M PGF 2␣ for 5 min causes a significant dose-dependent reduction in InsP formation as compared with control cells (Fig. 7A), although membrane receptor protein (Fig. 7B) and whole cell binding (408.8 dpm/10 6 cells in control group and 419.6 dpm/10 6 cells in the pretreated group) remain essentially unchanged. The dose response for pretreatment revealed that attenuation of InsP formation was maximal when cells were pretreated with ϳ1 M PGF 2␣ , resulting in ϳ60% desensitization (Fig. 7C). The ability of 12-iso-PGF 2␣ to cause FP desensitization was also tested on HEK-FP cells. When pretreated with PGF 2␣ or 12-iso-PGF 2␣ , FP undergoes rapid desensitization to both PGF 2␣ and 12-iso-PGF 2␣ (Fig. 8).
Both PGF 2␣ and 12-iso-PGF 2␣ induce InsP formation in NIH 3T3 cells in a dose-dependent manner (Fig. 9A). The EC 50 for InsP formation by PGF 2␣ is 50 Ϯ 8.3 nM. This is comparable to the EC 50 of PGF 2␣ for InsP formation in HEK-FP cells (Fig. 2) and also comparable to the EC 50 for InsP formation reported by Nakao et al. (47) in NIH 3T3 cells (ϳ46 nM). In NIH 3T3 cells, PGF 2␣ causes a dose-dependent increase in mitogenesis, with an EC 50 of ϳ25 Ϯ 3.8 nM (Fig. 9B). This response was also mimicked by 12-iso-PGF 2␣ . The mitogenic response parallels InsP formation, resulting in a maximum of ϳ3.2-fold increase over basal values. DISCUSSION We have cloned the FP from a hcb cDNA library, a likely target tissue for the efficacy of FP agonists in the treatment of glaucoma. Although the hcb-FP is identical to that cloned from the human uterus, the isolation of only two clones from the ocular source suggests that the FP is not expressed abundantly in the ciliary body. However, it does not rule out the existence of other FP isoforms (23) in other parts of the eye. Studies on the distribution of FP over a wide range of human tissues reveal its mRNA to be abundant in the human heart, in addition to reproductive tissues, as reported previously (19 -21). This is particularly interesting in light of recent reports on the ability of PGF 2␣ to cause hypertrophy of cardiac myocytes (12). Generation of HEK cells stably expressing the FP presents a tool for the detailed molecular characterization of FP. This is of importance, because there is a discrepancy between the rank order of potency of PGF 2␣ analogs in their ability to reduce IOP and their ability to bind FP in various membrane preparations (48). As with other prostanoid G protein-coupled receptors (45), the pattern of fluorescence observed with an FP-specific antibody suggests that FP is localized predominantly at the cell membrane. Availability of a human FP-specific antibody will facilitate determination of the pattern of FP receptor expression in the eye.
The molecular mechanisms of agonist-induced rapid FP receptor desensitization have not been elucidated to date. FPs are down-regulated in astrocytes after prolonged (Ͼ4 h) exposure to PGF 2␣ (49), and constriction of bovine sphincter muscle evoked by PGF 2␣ is down-regulated upon pretreatment of the preparation with the eicosanoid for 45 min (50). We now demonstrate that stimulation of HEK cells expressing FP, or of NIH 3T3 cells expressing endogenous FP, with PGF 2␣ results in rapid desensitization, initially without loss of receptor protein from the cell surface. The availability of these reagents is likely to facilitate investigation of the mechanism of action of PGF 2␣ analogs in ocular disease and of tachyphylaxis to FP agonists in the treatment of glaucoma. F 2 isoprostanes are free radical-catalyzed products of arachidonic acid (28). Up to 64 different isomers may be formed theoretically, belonging to four structural classes (41)(42)(43). Initially, these compounds are formed in situ on the cell membrane, from which they may be cleaved by the action of phospholipases to circulate and, ultimately, be excreted in urine (28). Specific measurement of isoprostanes in affected tissues, circulating lipoproteins, and urine holds promise as an approach to study oxidative stress in vivo. A more controversial issue is whether F 2 isoprostanes, or indeed analogous isomeric forms of other eicosanoids (29), might mediate some of the functional consequences of free radical generation. It has been speculated that in their esterified form, they may contribute to free radical-catalyzed membrane injury (28).
The biological effects of isoprostanes have only recently been investigated. Much attention has been paid to one member of the class IV F 2 isoprostanes, 8-iso-PGF 2␣ . This has been shown to stimulate inositol phosphate formation and DNA synthesis in cultured rat aortic smooth muscle cells (32). It is also a potent vasoconstrictor, at least in the renal and pulmonary circulations (51). It also stimulates mitogenesis and modulates platelet function, facilitating aggregation by subthreshold concentrations of conventional platelet agonists, such as ADP and thrombin (33). These effects of 8-iso-PGF 2␣ are blocked by  6. A, immunoblot analysis of FP. 10 g of membrane protein from HEK-FP cells, NIH 3T3 cells, or HEK 293 cells were electrophoresed on a 10% SDS-polyacrylamide gel, transferred to a nitrocellulose membrane, and probed with the anti-FP antisera as described under "Experimental Procedures." In some lanes, the anti-FP antisera were preincubated with 10 g/ml of the corresponding peptide. B, immunocytochemistry of FP. HEK-FP cells (top and bottom right) or HEK cells (bottom left) were fixed on chamber slides with 70% methanol and 30% acetone as described under "Experimental Procedures." Cells were incubated with anti-FP antibody with (bottom right) or without the corresponding peptide, and immunoreactivity was detected using a anti-rabbit fluorescein isothiocyanate secondary antibody under a fluorescence microscope (magnification, ϫ40).
pharmacological TP antagonists. However, the concentration of 8-iso-PGF 2␣ needed to evoke these effects seem much greater than that which circulates in vivo (33). Furthermore, 8-iso-PGF 2␣ , unlike other isoprostanes, may also be formed by a cyclooxygenase-dependent pathway (52).
We have recently synthesized several F 2 isoprostane isomers (43,(53)(54)(55). One of these, 12-iso-PGF 2␣ (43), activates the FP in a specific and saturable manner. It seems likely that 12-iso-PGF 2␣ may be an abundant member of the F 2 isoprostane family, inasmuch as free radical cyclization rules predict that upon formation of a cyclopentane ring, after oxidative modification of arachidonic acid, the adjacent substituents formed are cis to each other. Thus, cyclization of the hydroperoxy radical derived from 11-hydroperoxyeicosatetraenoic acid would lead predominantly to the formation of cis products such as 8-iso-PGF 2␣ and 12-iso-PGF 2␣ . Two reports actually predict the formation of 12-iso-PGF 2␣ type products in larger amounts than 8-iso-PGF 2␣ , as a result of such free radical cyclization (56,57).
Clearly, discrete isoprostanes might activate their own specific receptors. However, despite much speculation, no such receptors have been cloned to date and, save for the case of 8-iso-PGF 2␣ (which may also be formed enzymatically), specific receptors for the by-products of lipid peroxidation may seem unlikely. A more plausible concept is that isoprostanes act, in concert, as incidental ligands at prostanoid receptors. However, the comparative dose-response relationships for individual isoprostanes versus the natural prostanoid ligand, as exemplified in this report, reveals that highly concentrated forms of iso- Cells were then washed with 50 mM glycine, 150 mM NaCl, pH 3.0, and restimulated with varying concentrations of PGF 2␣ for 10 min at 37°C. Total InsP formation was measured as described under "Experimental Procedures." B, immunocytochemistry of HEK-FP cells pretreated with PGF 2␣ . HEK-FP cells plated in chamber slides were treated with vehicle (top) or 1 M PGF 2␣ (bottom) for 5 min at 37°C. Cells were then washed with PBS and processed for immunostaining as described under "Experimental Procedures." C, dose response for pretreatment. Confluent cultures of HEK-FP cells in 12-well plates were labeled to equilibrium with myo-[2-3 H]inositol (2 Ci/ml) for 16 -24 h in serum-free DMEM. Cells were treated with 20 mM LiCl for 15 min at 37°C and then pretreated with various concentrations of PGF 2␣ for 5 min at 37°C. Cells were then washed with 50 mM glycine, 150 mM NaCl, pH 3.0, and restimulated with 100 nM PGF 2␣ for 10 min at 37°C. Total InsP formation was measured as described under "Experimental Procedures." prostane delivery would be required for membrane receptor activation. Given the coordinate formation of multiple isomeric species, other considerations may pertain. For example, multiple isoprostanes might activate distinct eicosanoid receptors, which culminate in a common biological response. To address this possibility, we explored the capability of two structurally distinct members of class IV isoprostanes, 8-iso-PGF 2␣ and 12-iso-PGF 2␣ , to activate the FP and TP isoforms, distinct receptors which mediate common biological responses, such as vasoconstriction and mitogenesis. 12-iso-PGF 2␣ and 8-iso-PGF 2␣ activate the FP and TP isoforms, respectively. However, neither compound activated the IP, which mediates vasodilation. Thus, F 2 isoprostanes may act cooperatively to facilitate a common biological response via distinct eicosanoid receptors. Isoprostanes may also desensitize the response of eicosanoid receptors to their natural ligand. We have shown previously that 8-iso-PGF 2␣ may cross-desensitize TPs. Similarly, we now demonstrate that 12-iso-PGF 2␣ may cross-desensitize the human FP.
In summary, we have cloned a human ocular FP, generated HEK 293 cells stably expressing the ocular FP, and demonstrated membrane localization of the FP protein in these cells. The availability of these reagents will facilitate investigations into the molecular basis of action of PGF 2␣ analogs in reducing IOP. Furthermore, we have demonstrated that the FP may be activated and desensitized, not only by its natural ligand, PGF 2␣ , but also by F 2 isoprostanes like 12-iso-PGF 2␣ . These observations raise the possibility that the therapeutic response to PGF 2␣ analogs may be modulated by F 2 isoprostanes in syndromes of oxidant stress, such as glaucoma or congestive heart failure. FIG. 9. A, inositol phosphate formation in NIH 3T3 cells. Confluent cultures of NIH 3T3 cells in 12-well plates were labeled to equilibrium with myo-[2-3 H]inositol (2 Ci/ml) for 16 -24 h in serum-free DMEM. Cells were treated with 20 mM LiCl for 15 min at 37°C and then stimulated with various concentrations of PGF 2␣ or 12-iso-PGF 2␣ for 10 min at 37°C. Total InsP formation was measured as described under "Experimental Procedures." B, mitogenesis in NIH 3T3 cells. Confluent quiescent cultures of NIH 3T3 cells in 12-well plates were stimulated in serum-free DMEM for 24 h at 37°C with varying concentrations of PGF 2␣ or 12-iso-PGF 2␣ . In the last 2 h of incubation, 0.5 Ci/ml [ 3 H]thymidine was added to the medium. The reactions were stopped, and thymidine incorporation was measured as described under "Experimental Procedures."